Unit Region Encoding: A Unified and Compact Geometry-aware Representation for Floorplan Applications

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Huichao, Wang, Pengyu, Li, Manyi, Li, Zuojun, Wu, Yaguang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917897059696640
author Zhang, Huichao
Wang, Pengyu
Li, Manyi
Li, Zuojun
Wu, Yaguang
author_facet Zhang, Huichao
Wang, Pengyu
Li, Manyi
Li, Zuojun
Wu, Yaguang
contents We present the Unit Region Encoding of floorplans, which is a unified and compact geometry-aware encoding representation for various applications, ranging from interior space planning, floorplan metric learning to floorplan generation tasks. The floorplans are represented as the latent encodings on a set of boundary-adaptive unit region partition based on the clustering of the proposed geometry-aware density map. The latent encodings are extracted by a trained network (URE-Net) from the input dense density map and other available semantic maps. Compared to the over-segmented rasterized images and the room-level graph structures, our representation can be flexibly adapted to different applications with the sliced unit regions while achieving higher accuracy performance and better visual quality. We conduct a variety of experiments and compare to the state-of-the-art methods on the aforementioned applications to validate the superiority of our representation, as well as extensive ablation studies to demonstrate the effect of our slicing choices.
format Preprint
id arxiv_https___arxiv_org_abs_2501_11097
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unit Region Encoding: A Unified and Compact Geometry-aware Representation for Floorplan Applications
Zhang, Huichao
Wang, Pengyu
Li, Manyi
Li, Zuojun
Wu, Yaguang
Computer Vision and Pattern Recognition
We present the Unit Region Encoding of floorplans, which is a unified and compact geometry-aware encoding representation for various applications, ranging from interior space planning, floorplan metric learning to floorplan generation tasks. The floorplans are represented as the latent encodings on a set of boundary-adaptive unit region partition based on the clustering of the proposed geometry-aware density map. The latent encodings are extracted by a trained network (URE-Net) from the input dense density map and other available semantic maps. Compared to the over-segmented rasterized images and the room-level graph structures, our representation can be flexibly adapted to different applications with the sliced unit regions while achieving higher accuracy performance and better visual quality. We conduct a variety of experiments and compare to the state-of-the-art methods on the aforementioned applications to validate the superiority of our representation, as well as extensive ablation studies to demonstrate the effect of our slicing choices.
title Unit Region Encoding: A Unified and Compact Geometry-aware Representation for Floorplan Applications
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.11097